The document describes an IoT-based smart black box system for vehicles. The system uses various sensors like speed sensors, alcohol sensors, and seat belt sensors to continuously monitor vehicle parameters. If an accident occurs, sensor data is sent via IoT to a mobile app to help determine insurance claims. The system aims to improve vehicle and driver safety by predicting risks, monitoring driver behavior, and enabling remote health monitoring of vehicles. It allows for preventative maintenance and reduces costs from accidents, fines, and unnecessary repairs. The system provides real-time monitoring and alerts to improve response time to issues and enhance the customer experience.

1. International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 06 Issue: 05 | May 2019 www.irjet.net p-ISSN: 2395-0072
© 2019, IRJET | Impact Factor value: 7.211 | ISO 9001:2008 Certified Journal | Page 1512
IoT based smart BLACK BOX System
Aishwarya Bhadale1, Chetna Kalbhor2, Rishikesh Khaladkar3
Professor: Pratiksha Patil, Electronics and Telecommunication Engineering, Keystone school of Engineering,
Maharashtra, India
---------------------------------------------------------------------***----------------------------------------------------------------------
Abstract -The objective of this project is safetyandsecurity
of vehicle user and get the insurance. The main aim of project
to avoid accident using speed sensor, alcohol sensor, patti
switch. speed sensor calculate the speed at which the wheels
are rotating. This ultimately determines the speed at which
vehicle is traveling. Speed sensor sends the speed information
to the control . So that can regulate other function of vehicle.
Alcohol sensor check the how much alcohol consume by the
vehicle driver. Patti switch indicatesthatdriverwearinga seat
belt or not. Depending upon technology is implemented, the
vehicle driver will get the notification on his app. If he an
accident the data will be send through IOT to the app on
mobile for an insurance. When sensor data is monitored and
mining is performed it would be possible to predict the good
health of vehicle.
If we turn on GPS system then it sends a vehicle location
through a message on mobile. Using this we can get vehicle
location.
Key Words: PIC microcontroller, Gas sensor, Alcohol
sensor, Patti switch sensor.
1. INTRODUCTION
An accident is a unpleasant event that no one ever want to
occur in their life. .It ruins the life of many people. The
percentage of accident raised the eyebrows and caught the
attention. It found that 80% of time is the fault of driver. So
avoid the accident in today’s world science has made
amazing advances in a car. These cars are more advanced
than ever.
We have design such a system in which all the parameters
which are responsible for accident will get recorded & help
us to prevent the accident and according to recorded data
the vehicle driver will get insurance.
The project is developed to record the informational data
such as alcohol level, speed, gas leakage and driver will
wearing a seat belt or not etc. It can also use for vehicle
mapping & alert with the help of GPS technology,
This system utilize the sensor data of car& predicts the
health of car. The utilized data provides on local app which
alerts the driver. If accident is occured IOT through the
reason for accident is provided to the insurance company to
get the insurance. According to the reason of accident the
driver will get the insurance amount. If we turn on the GPS
system it continuously provide the car location through
message. When accident occurred it send the message that
accident is occurred.
1.1 Literature Survey
The literature surveyed some different papers to get
information about the existing work which have been done
by T.Gayathri, Kargupta, Hillol
Vehicle health monitoring system :Vehicle health
monitoring utilizes the sensor on data vehicle, minethedata
& predicts the health of car. It provide an idea on when the
vehicle would need maintenance. The model also can warn
driver dozes off or in case of drunkendriving.The purposeof
this research paper is to give an overview of the existing
vehicle health monitoring system. It gives the different ways
in which vehicle health can be monitored. Vehicle
monitoring can be distributed or on-board data mining.
Computation or mining in vehicle are restricted by less
memory and processor capacity, still an on board mining is
more advantageous than fully distributed data mining
system. This paper survey the existing model for efficient on
board vehicle health monitoring system.
Uferah Shafi, Asad Safi have following informationof Vehicle
Remote Health Monitoring :In many industries inclusive of
automotive vehicle industry, predictive maintenance has
become more important. It is very difficult to diagnose
failure in advance in the vehicle industry because the
availability is limited of sensors and some of the designing
exertions. An approach is presented in this article for fault
reduction of four main subsystems of vehicle, fuel system,
ignition system, exhaust system, and cooling system. When
vehicle is on the move the sensor collected, both in faulty
condition (when any failure in specific system hasoccurred)
and in normal condition.
HARMAN Ignite Platform(2015)Vehicle health hand
diagnostic dashboard: Before the advent of connected
vehicles , consumers simply drove their new vehicle out of
deals reacted to issue as they arose. After combining
diagnostics and connectivity, managers and service
providers are able to offer additional services to their
customers. Ecosystem provide contextual and advanced
services like early field warnings, predictive maintenance,
remote diagnostics and repair, fuel management. This
reduces operations costs of the vehicle, reaction time to
issues, accident liability and fines and also enhances
customer experience.

2. International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 06 Issue: 05 | May 2019 www.irjet.net p-ISSN: 2395-0072
© 2019, IRJET | Impact Factor value: 7.211 | ISO 9001:2008 Certified Journal | Page 1513
VEDAS(1) works on the concept of ubiquitous data mining.
An in device , real time mining of data on a ubiquitous
computing environment in accordance to the environment
requirements by considering resource constraints of the
device, exploiting context the information behaving
autonomously, applying special privacypreservingmethods
in VEDAS and mine fleet, partial on board data mining is
done.
1.2 Block Diagram and discription
Fig -1: Block diagram of proposed system
In the proposed system we aredesigninga system whichwill
continuously monitor the vehicle parametersandsendthem
on Android APP on internet of things (IOT). All the sensors
are interfaced to inbuilt ADC 10 bit. The µC will scan all the
parameters and send them on Android APP using MQTT
protocol.
1) PIC MICROCONTROLLER:
Fig -2: PIC microcontroller
PIC18F4520-I/P isa40-pinFlash-
based32KBmicrocontrollerwith13x10-bitA/D
Converter,2xcomparators,UART,A/E/USART,SPI,I2C1-
MSSP(SPI/I2C) and ECCP peripherals.
Specification:
CPU
 Upto10MIPSPerformanceat3V
 Compiler optimized RISC architecture
 8x8SingleCycleHardwareMultiply
System
 Internaloscillatorsupport-31kHzto8MHzwith4xPLL
 Fail-Safe Clock Monitor-allows safe shutdown if
clock fails
 Watchdog Timer with separate RC oscillator
 WideoperatingVoltagerange;2.0Vto5.5VnanoWattP
owerManagedModes
 Run, Idle and SLEEP modes
 Idlemodecurrentsdownto5.8uAtypical
 Sleepmodecurrentsdownto0.1uAtypical
Analog Features
 10-bitADC,13channels,100Ksamples per second
 Programmable Low Voltage Detection Module
 Programmable Brown-out-Reset Module
 Two Analog Comparators multiplexing Peripherals
 Master Synchronous Serial Port supports
 SPI™andI2C™masterandslavemode
 EUSART module including LIN bus support
 Four Timer modules
 Upto5PWMoutputs
Microcontroller
PIC 18F
Alcohol
CNG
sensor
Seat belt
Speed
Sensor
Accident
sensor
LCD
Buzzer
Wi-Fi
Modem
Relay
Ignition
Android App
(Local App)
Android App
(IOT App)

3. International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 06 Issue: 05 | May 2019 www.irjet.net p-ISSN: 2395-0072
© 2019, IRJET | Impact Factor value: 7.211 | ISO 9001:2008 Certified Journal | Page 1514
2)ESP:
Fig -3: ESP
Espressif Systems’ SmartConnectivity Platform (ESCP)
enables sophisticated features including:
 Fast switch between sleep and wakeup mode for
energy-efficient purpose
 Adaptive radio biasing for low-power operation
 Advance signal processing
 Spur cancellation and RF co-existence mechanisms
for common cellular, Bluetooth,DDR, LVDS, LCD
interference mitigation.
Specifications:
 Operating Voltage 2.5V ~ 3.6V
 Package Size QFN32-pin (5 mm x 5 mm)
 Operating Current Average value: 80 mA
 Operating Temperature Range –40°

3) MQTT PROTOCOL
This is simply made for the connections with the remote
locations. It is an ISO standard and publish-subscribe-based
messaging protocol. It works on top of the TCP/IP protocol.
Whenever an accident is detected the µC will send all the
data to local APP which will further send the data to IOT
based server. On the android APP we have a graphical user
interface to display the data. The data is analyzed and then
an appropriate notification is displayed on APP, and if the
accident is detected then the message will send on the
registered mobile numbers. Depending upon the data the
APP will continuously monitor the data pattern and as and
when any parameter crosses the setpointtheAPP will either
display or send a notification to user to take appropriate
action so that the vehicle / Engine health is maintained.
The main benefits of vehicle health monitoring
1. Reduces unscheduled vehicle downtime through
automated remote diagnostics and alerts
2. Just-in-time vehicle maintenance’ saves money on
unnecessary maintenance
3. Reduces the risk of accidents, fuel expenses, insurance
liability and fines by effective driver behaviour
monitoring
4. It integrates with existing backend solutions facilities
OTA updates and remote diagnostics.
5. Helps in deploying new services and generating
aftermarket service revenue opportunities
6. Improved management of overtime andcostaccounting
of jobs through regular review of trip data and overtime
dashboard report
Table -1: Comparison with existing techniques
S
R
N
O
Authors Paper Advantages
1
.
T.Gayatri Vehicle Health
monitoring And
prediction
system
Real time vehicle
monitoring
2
.
Kargupta,
Hillol, et.al.
A distributed
data streaming
and mobile
system for real
time vehicle
monitoring
system
Vehicle failure
prediction
2. CONCLUSIONS
With this increasing graph of using smart phones and
wireless communication, it is become easy to use this
technologies for real time solutions.
Despite of the limited resources, these technologies are
being used along with machine learning approachestosolve
big problems in automotive industry. Our first contribution
to this paper is data generation and feature selection.Sensor
data of many car manufactures has been used. WE have
selected those sensors which can cause a system to break
down. And the second contribution is to built a user friendly
vehicle fault prediction, vehicle remote health monitoring
system.
REFERENCES
[1] R.Mangharam,J.Meyers,R.Eakumar,D.Stancil,J.Parikh,H.
“A multi hop mobile networking test bed for telematics”
society for Automotive engineers world congress
Detroit, USA. April 2005.
[2] DARPA. The DARPA urban challenge,
2007.http://www.darpa.mil/grandchallenge.
[3] Kurt Dresner & Peter Stone (2008), Replacing the Stop
Sign: Unmanaged Intersection Control, The Fifth

4. International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 06 Issue: 05 | May 2019 www.irjet.net p-ISSN: 2395-0072
© 2019, IRJET | Impact Factor value: 7.211 | ISO 9001:2008 Certified Journal | Page 1515
Workshop on Agents in Traffic andTransportationMulti
agent Systems. pp. 94-101, Estoril, Portugal.
[4] Kargupta, Hillol, et al. "A mobile and distributed data
stream mining system for real-timevehiclemonitoring."
Proceedings of the 2004 SIAM International Conference
on Data Mining. Society for Industrial and Applied
Mathematics, 2004.
[5] Kargupta, Hillol, et alNew Generation Computing 25.1
(2006): 5-32.
[6] Kargupta, Hillol. "Onboard vehicle datamining social
networking, advertisement." U.S. PatentNO.8,478,514.2
JUL.2013
[7] Stewart, Gilbert W. "Matrix perturbationtheory” (1990)
[8] Basilevsky. Statistical factor analysis and related
Methods:theory and application ,1994.